added file permissions and fixed a duplicate output bug.

app/services/streaming.py

@@ -4,11 +4,11 @@ from collections.abc import AsyncIterator
 
 from rich.live import Live
 from rich.markdown import Markdown
+from rich.panel import Panel
 
 from app.models.config import DisplayConfig
 from app.models.message import Message
 from app.models.tool_call import ToolCall, ToolCallFunction
-from app.utils.display import print_assistant_message
 from app.utils.logging import console, get_logger
 from app.utils.token_counter import TokenUsage
 
@@ -50,14 +50,19 @@ class StreamHandler:
                 # Show reasoning while waiting for content
                 display_text = self._accumulated_content
                 if not display_text and self._accumulated_reasoning:
-                    display_text = f"*thinking...*"
+                    display_text = "*thinking...*"
 
                 if display_text and self._display_config.stream_output:
-                    live.update(Markdown(display_text))
-
-        # Final static render
-        if self._accumulated_content:
-            print_assistant_message(self._accumulated_content)
+                    # Render inside the same Assistant panel used for final output
+                    # so the live display and final frame are visually consistent
+                    live.update(
+                        Panel(
+                            Markdown(display_text),
+                            title="Assistant",
+                            border_style="green",
+                            expand=True,
+                        )
+                    )
 
         tool_calls = self._build_tool_calls() or None
         return Message(